AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

SAEs-BrainMap: Unveiling the Emergence of Specialized Concepts in Deep Models via Brain Alignment

Ziming Mao, Jia Xu, Wenxuan Pan, Mufan Xue, Yaochu Jin, Guoyuan Yang

ICML 2026 · Vision-language and multimodal › Human and embodied modeling · approach: fMRI visual decoding

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use CLS-trained SAE features instead of raw neurons (ablation)ViT-B/16 CLIP · NSD S5max cosine similarity: 0.5875 -> 0.7247Improved
use CLS-trained SAE features instead of raw neurons (ablation)ViT-B/16 CLIP · NSD S5RSA score: 0.511 -> 0.580Improved
use patch-trained SAE features instead of raw neurons (ablation)DINOv2 · NSD S5max cosine similarity: 0.6961 -> 0.6517Hurt
use patch-trained SAE features instead of raw neurons (ablation)ImageNet ViT · NSD S5RSA score: 0.596 -> 0.656No clear effect
select features with brain fMRI guidance instead of CLIP text matching (CLIP-Dissect, CLIP-SAEs) (ablation)ViT-B/16 CLIP · face-selective featureslayer of emergence and semantic coherenceImproved
train SAEs on all layers to trace concept emergenceViT-B/16 CLIP · Face conceptemergence layer: emerges at Layer 2Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy